1. Field of the Invention
The present invention relates to a data search method and a data search system using thereof which conduct searching in an electronic document database or the like according to an index and in particular relates to a data search method and a data search system using thereof which output the result of searching an d display it in a mode showing its features.
2. Discussion of the Related Art
Databases related to the academic works, publications, newspapers or the like have various types such as an index database recording only the bibliographic items, an abstract database recording abstracts in addition to the bibliographic items, and a full text database recording the whole texts. Recently, the full text database has been widely used because of great store of information. Such state of using database is described in "Journal of Information Processing Society of Japan", Vol. 33, No. 4, April 1992, pp. 413-420 and the same, Vol. 33, No. 10, October 1992, pp. 1144-1153.
In searching in formation recorded in such database, the searching process can be quickly carried out by utilizing an index using a word (a string) contained in the information as a key. In particular, it can be said that it is indispensable for realizing search in the full text database in a practical process time to conduct searching according to the index.
Generally, the index describes a string (word) occurring in the document, and identification informations of the document containing the strings or identification informations of sub-documents constituting a document. As a result of searching in the plural documents (a group of documents), the identification informations of the documents containing the strings inputted for searching (for example, the names of the documents) are listed, and on the other hand, as a result of searching in a single document, the identification informations of the sub-documents containing the strings inputted for searching (for example, the page numbers) are listed, whereby the user is able to know which is the document or the sub-document containing the string inputted for searching (namely, the string as a searching index). Regarding the listed identification informations of the sub-documents, the content of text of the corresponding sub-document can be displayed in accordance with the user's instruction.
The frequency of occurrence of the string to be searched (in some cases, containing its synonyms) in each of the documents is helpful information for determining which document is optimum as a result of searching among the documents listed with their document names. Therefore, the frequency of occurrence is numerically displayed by the side of each of the listed document names, or the frequency of occurrence is displayed by a graph with the list of the document names.
As described above, in the search system simply displaying the identification informations of the documents or sub-documents in the list, it is difficult to determine for the user which document is optimum as the result of searching based on the display. In the case where the content of the text corresponding to the listed sub-document is displayed, determination of optimumness can be easily made based on the display. However, in this case, each and every content of the text must be displayed corresponding to each of the identification informations of a large number of listed sub-documents, which results in complex operations and a long process time.
In the case of attaching the frequency of occurrence of the searched string to the list, determination of optimumness is easily made with reference to the frequency of occurrence.
However, if the numerals indicating frequency of occurrence is displayed by the sides of many listed documents, it is difficult for the user to visually recognize each value of frequency of occurrence, and besides, it is not very much effective in practice because a large number of numeric values indicating the frequency of occurrence must be compared with each other. In the case where the graph showing the frequencies of occurrence is displayed with the list of the document names, variation in frequencies of occurrence can be recognized at a glance. However, the space for displaying graph is required in the screen and thereby the space for the list of the document names which are most important is reduced, thus making it harder for the user to recognize the list.